QRCode

• Numbers
• Alphanumeric Characters
• Bytes (8-bit Latin-1 characters)
• Kanji

Each version of QR codes come in a variety of different sizes. For instance, version 1 is 21 x 21 pixels and version 40 is 177x177. In order to find the size needed for a QR code (based by version), we simply multiply the version by 4, then add 17. (version * 4) + 17.

Each QR code contains error correction blocks. There are four different levels of error correction.

This means that if part of the QR code is covered, the code can still be read. With a higher correction level, more of the QR code can be covered. Some like to put a company logo in the center and use a higher correction level for advertising.

Every QR code has the same pattern. QR codes are a sqaure made up of dark and light "boxes" and each box is in a specific sequence.

• Finder Pattern: The finder pattern is made up of three 7x7 squares at the top left, top right, and bottom left corners.
• Separators: The separators are located around the finder pattern. This separates the finder pattern from the rest of the QR code and they are always made up of light colored blocks.
• Timing Pattern: The timing pattern is made up of alternating light and dark blocks, this allows the decoder to determine the width of a single block.
• Alignment Pattern: The alignment patterns are made up of a 5x5 square. There are n^2 - 3 (n being height or width) alignment patterns in a QR code unless it is version 1. The alignment pattern will not be placed where a finder pattern is located.
• Format Information: These are blocks stored next to the finder patterns that store information about the error correction level and encoding mode.
• Version Inforation: The version information is made up of two 6x3 blocks located to the left of the right finder block, and above the bottom left finder block. This will be found on versions 7+.
• One Dark Block: One dark block is placed in the 9th column, and in the ((4 * version) + 10) row.

qr-parts

1. Determine the encoding method used based on given text.
2. Determine the version and error correction (if none is given).
3. Convert the encoding mode to bits.
4. Convert the number of code words to bits.
5. Convert each character to bits (For numeric and alphanumeric modes the characters need to be grouped before being encoded). The amount of bits needed per character also needs to be determined.
6. Add terminator block if possible.
7. Add padding bytes until the capacity is reached.
8. Draw all standard QR patterns.
9. Draw all encoded codewords in a zig zag pattern.
10. Apply mask and determine which mask has the lowest penalty score.